CN107633052A - Intelligent robot manufacture method - Google Patents
Intelligent robot manufacture method Download PDFInfo
- Publication number
- CN107633052A CN107633052A CN201710846773.5A CN201710846773A CN107633052A CN 107633052 A CN107633052 A CN 107633052A CN 201710846773 A CN201710846773 A CN 201710846773A CN 107633052 A CN107633052 A CN 107633052A
- Authority
- CN
- China
- Prior art keywords
- data
- tree
- shaped
- database
- storehouse
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 23
- 238000004458 analytical method Methods 0.000 claims abstract description 114
- 238000003860 storage Methods 0.000 claims abstract description 22
- 230000007935 neutral effect Effects 0.000 claims abstract description 5
- 239000013589 supplement Substances 0.000 claims description 13
- 230000007812 deficiency Effects 0.000 claims description 12
- 230000033001 locomotion Effects 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 10
- 239000000047 product Substances 0.000 claims description 2
- 235000012054 meals Nutrition 0.000 description 36
- 238000003672 processing method Methods 0.000 description 7
- 238000012937 correction Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 241001269238 Data Species 0.000 description 4
- 208000027418 Wounds and injury Diseases 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000006378 damage Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 208000014674 injury Diseases 0.000 description 2
- 210000002784 stomach Anatomy 0.000 description 2
- 244000144730 Amygdalus persica Species 0.000 description 1
- 241000079786 Erotesis Species 0.000 description 1
- 240000007711 Peperomia pellucida Species 0.000 description 1
- 235000012364 Peperomia pellucida Nutrition 0.000 description 1
- 235000006040 Prunus persica var persica Nutrition 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000003466 anti-cipated effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 235000013311 vegetables Nutrition 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Intelligent robot manufacture method.The present invention relates to a kind of intelligent robot manufacture method.Step 1：From input window input sentence give first sky file store carry out storage as session log and meanwhile also give tree-shaped logic analysis judge database analyzed and judge then by the result analyzed and judged export completion to output window export and meanwhile the result analyzed and judged can also be given sky file store carry out store as session log；Step 2：Tree-shaped logic analysis judges that database can need to read the data of tree-shaped thinking framework lane database or to tree-shaped thinking framework data database storing in analysis and judgement according to data；Step 3：Tree-shaped logic analysis judges that database can need to read empty file store data or to neutral gear case library storage data in analysis and judgement according to data.The present invention is used for intelligent robot manufacture method.
Description
Technical field
The present invention relates to a kind of intelligent robot manufacture method.
Background technology
There is very large deviation in present artificial intelligence, the problem of answering is in the presence of mistake in logic to the understanding of instruction.
The content of the invention
It is an object of the invention to provide a kind of intelligent robot manufacture method, is not in patrol to solve the above problems
Mistake on volume, makes communication more smooth.
Above-mentioned purpose is realized by following technical scheme：
A kind of intelligent robot manufacture method, is realized by following steps：
Step 1：Robot oneself first using the sentence inputted from input window give first sky file store carry out storage as pair
Also give tree-shaped logic analysis judges that database is analyzed and judges that data can be read in analysis and judgement to words record simultaneously
Supplement storehouse in data assist analysis then by the result analyzed and judged export to output window complete output while
The result analyzed and judged can be given to sky file store and carry out storage as session log；
Step 2：Tree-shaped logic analysis judges that database can need to read tree-shaped thinking framework in analysis and judgement according to data
The data of lane database assist analysis or to tree-shaped thinking framework data database storings；
Step 3：Tree-shaped logic analysis judges that database can need to read empty file store data in analysis and judgement according to data
Assist analysis or to neutral gear case library storage data；
Step 4：When tree-shaped logic analysis judges database, tree-shaped thinking framework database, read during empty file store data deficiencies
Data in data filling storehouse also can judge database to the stock as supplementary data or as needed from tree-shaped logic analysis
Store up data.
Described intelligent robot manufacture method, described tree-shaped logic analysis judge that the data of database are all tree-shaped
And be that the sentence of same meaning or word, nearly adopted sentence or word are put together optional one as output data, but it is necessary to export this data
Energy and empty file store, tree-shaped thinking framework database, data filling storehouse and tree-shaped logic analysis judge other numbers of lane database
According to that can match, tree-shaped logic analysis judges that the data of database are exactly how many kind implication of a word and the words is needed to divide
For the data of analysis all by behind the coming of row one-level one-level, tree-shaped logic analysis judges that database has numerous such data group
Into tree-shaped logic analysis judges that the data of lane database should be included to semantic analysis data and to cycle, volume, speed, weight
Amount, volume, volume, pressure, inertia force, area, height, interpersonal relation, the relation between object and object, motion
Rule, movement technique, object structures, the processing data of mobile changing rule method etc. between position and position.
Described intelligent robot manufacture method, the data of described tree-shaped thinking framework database are all tree-shaped, trees
The data of shape thinking framework database be all by behind a word how many kind answer method all by row, be listed in this knot below
It is all that synonym or sentence, near synonym or sentence are put together optional one as defeated that structure, which is also referred to as context mechanism and every words,
Go out data but export this data to must be able to judge database with empty file store, tree-shaped logic analysis, data filling storehouse and tree-shaped
Other data of thinking framework lane database can be matched, and tree-shaped thinking framework database is made up of numerous such data
（The word that the data of tree-shaped thinking framework lane database hereafter have, word or sentence be accordingly to be regarded as can with the word that associates above, word or sentence,
Otherwise hereafter without word, word or sentence are accordingly to be regarded as can not be with the word that associates above, word or sentence）.
Described intelligent robot manufacture method, the data of described empty file store are made up of two parts, this two parts number
According to being dummy status, empty file store is exactly the memory system of robot, and the data of empty file store divide two parts, and a portion is
Record includes analyzing the result exported after judgement on the information of robot oneself, and another part is the people that record is talked with it
Information includes the people that talks with it, and what is said or talked about.
Described intelligent robot manufacture method, data filling storehouse are divided into three parts：
Part I carries out supplementary data to empty file store and used, when empty file store data deficiencies standby it supplement；
Part II to tree-shaped thinking frame data frame storehouse carry out supplementary data use, when tree-shaped thinking frame data frame storehouse data not
When sufficient standby it supplement；
Part III judges that tree-shaped logic analysis database carries out supplementary data and used, when tree-shaped logic analysis judges data
During the data deficiencies of storehouse standby it supplement.
Beneficial effect：
1. the present invention can carry out smoothly language communication.
2. the present invention can the higher sentence of response.
3. the clear logic of the present invention, sentence is smooth.
4. the robot of the present invention has independent thinking ability.
5. present invention installation machine vision sensor etc. can drive a tank, the complete independently task such as fighter plane.
6. the present invention can be used as operational staff's fast reaction to assist fight, 24 hours standby.
7. the present invention can do personal assistant network customer service etc..
8. the present invention can learn professional technique.
9. the present invention can make educational robot can be to the man-to-man education of student.
10. the present invention, which can make be monitored spatial domain after radar robot, can be used in aircraft carrier airport to aircraft
Carry out rise and fall commander and management and control.
Embodiment：
Embodiment 1
A kind of intelligent robot manufacture method, is realized by following steps：
Step 1：Robot oneself first using the sentence inputted from input window give first sky file store carry out storage as pair
Words record also gives tree-shaped logic analysis and judges the knot that database is analyzed and judges and then will analyze and judge simultaneously
Fruit exports completes output while can also give the result analyzed and judged to sky file store to carry out storage work to output window
For session log；
Step 2：Tree-shaped logic analysis judges that database can need to read tree-shaped thinking framework in analysis and judgement according to data
The data of lane database assist analysis or to tree-shaped thinking framework data database storings；
Step 3：Tree-shaped logic analysis judges that database can need to read empty file store data in analysis and judgement according to data
Assist analysis or to neutral gear case library storage data；
Step 4：When tree-shaped logic analysis judges database, tree-shaped thinking framework database, read during empty file store data deficiencies
Data in data filling storehouse also can judge database to the stock as supplementary data or as needed from tree-shaped logic analysis
Store up data.
Embodiment 2
Intelligent robot manufacture method described in embodiment 1, described tree-shaped logic analysis judge that the data of database are all trees
Shape and also be that the sentence of same meaning, nearly adopted sentence are put together optional one as output data, but export this data must be able to and
Empty file store, tree-shaped thinking framework database, data filling storehouse and tree-shaped logic analysis judge other data energy of lane database
Enough matchings, tree-shaped logic analysis judge that the data of database are exactly how many kind implication of a word and the words is needed to analyze
Data all by behind the coming of row one-level one-level,（Data in this storehouse, which run into name, includes you, I, he, man, Ms etc.
If word all inserts the first person, the second person, the third person and the first person with bracket to be empty instead of still bracket the inside,
At present DBMS structure also differs the name of the name woman of the second person, third person sensing man woman or man
Sample）, it is to semantic analysis data and heavy to cycle, volume, speed that tree-shaped logic analysis judges that the data of lane database should include
Amount, volume, volume, pressure, inertia force, area, height, interpersonal relation, the relation between object and object, motion
Rule, movement technique, object structures, the processing data of mobile changing rule method between position and position.
Described tree-shaped logic analysis judges that the data of database are all tree-shaped, and tree-shaped logic analysis judges database
Data are exactly in short how many kind implication and to need analyze data all by behind the coming of row one-level one-level,（In this storehouse
The DBMS of data first is exactly that in short, corresponding data are exactly the analyze data to the words behind the words, in this storehouse
It is made up of numerous such data）.
(1) tree-shaped logic analysis judges the function of database：The data function in this storehouse is that input is relaxed to be easy to input
Match somebody with somebody and read statement is analyzed and judged（Including the analysis and judgement to logic）According to the needs of analysis judgement, it can read
Take sky file store, tree-shaped thinking frame data frame storehouse, data filling storehouse, inner data are analyzed and judge and then will analyze and sentence
Disconnected result is exported to output window（Tree-shaped logic analysis judges that database can also be sentenced according to the information difference of input by analysis
Have no progeny as needed respectively to empty file store, tree-shaped thinking framework database, data filling library storage data）.
(2) tree-shaped logic analysis judges the operation principle of database：This storehouse is that the sentence inputted from input window is entered
Then row matching is analyzed the data matched（Matching is exactly that a comparison process is from input window by robot oneself
The sentence of mouth input is analyzed after lane database faceted search to the data that can be coincide by corresponding data behind sentence）
（It by data come these analyze datas analyzed all is that preset in advance is good that the analysis of this storehouse, which is,）This storehouse can judge according to analysis
Need also read empty file store, tree-shaped thinking framework database, data filling storehouse, the inner data progress cycle judges interpersonal pass
The various judgements such as system's judgement（As people has a meal daily, 3 meals are all eaten daily, regular meal regular time, and passage time is just
It can judge whether you have arrived the time having a meal daily, be exactly to judge in the cycle to this judgement）, interpersonal relationships judgement（Such as to one
The personal time that how many understands and got along carries out whether comprehensive descision is which type of friend friend is（As we give
Robot specifications get along 10 days be regular friend, you get along with robot 10 days it you will be treated as regular friend pair
Treat）, what etc. can be between friend can do between what household, and the option that can also read friend inside empty file store is deposited
The data of storage directly judge that someone is your friend）, analysis（Analysis includes adjusting the distance, speed, height, volume, area
Calculating and the relation between them various judgements be not inconsistent if my a meal eats up a buffalo the words it is logical so
He needs to know whether people's a meal will eat up the volume of food and be calculated between the volume of thing to be eaten up proportional
She must also know that what can eat the judgement what cannot eat etc. and to the structure of laws of motion and object
Judgement etc.（What can eat the ability what cannot eat and to judgement of laws of motion and the structure of object etc.
Be by read the description inside data filling storehouse to word or word or short sentence come）,（If this storehouse analyze two with
On result one is selected with regard to whose machine, this storehouse must have math block be used for calculate, this storehouse to other storehouses write data when
Otherwise time must be also write can go out first run-time error）；
(3) the processing method of question sentence：The processing method of question sentence is that the asked content of reading is then right in the storage result of empty file store
The question sentence content of input is analyzed and judges and then analysis judged result is exported to output window（Such as ask that you have had a meal,
Read the result stored in the option of having a meal of empty file store and then judge to export the processing completed to question sentence by analyzing）；
(4) the processing method of such as sentence：Such as the processing method of sentence be the things reason to know and unknown things reason by
Sentence of same meaning processing, and the such as content of such as sentence is carried out to judge whether that result that is correct and then judging analysis is defeated by analyzing
Go out（It cannot filled after object as people has had enough to be at table to fill as same container, the thing that people eats
The stomach of people is all attached to, is exactly a container in the stomach of people so this is such as correct）；
(5) the processing method of declarative sentence：The processing method of declarative sentence is to give the those set forth of declarative sentence to sky file store by neutral gear
The data options in case storehouse store and then input declarative sentence are carried out logic analysis and judges to handle and then the knot after processing
Fruit exports（As I has just had a meal, processing method is the option having a meal first to be found in empty file store and then the option setting that will have a meal
Judging into the state eaten up and then by analysis output, the analysis determination methods of the words of just having been had a meal to me are exactly：You are
Who, I either with or without necessity tell you I just had a meal or eaten is all what）；
(6) tree-shaped logic analysis judges database characteristics：Data in this storehouse are that the closely adopted sentence of all sentences of same meaning is put together
Optional one as output data but exports this data and must be able to and empty file store, and tree-shaped thinking framework database, data are mended
Filling other data in storehouse and Ben Kuli can match,（Data in this storehouse run into name
If empty insert the first person, the second person, the third person and the first person, the second people with bracket instead of still bracket the inside
Claim, data structure is also different when the third person points to the name of the name woman of man woman or man）This storehouse
In data should include to semantic analysis data and to cycle, volume, speed, weight, volume, volume, pressure, inertia force, face
Product, height, interpersonal relation, the relation between object and object, the characteristics of motion, movement technique, object structures, point
The structure of sub- atom, the various data of mobile changing rule method etc. between position and position；
(7) analytics：Analytics is necessary have these logical datas of logical data be also it is tree-shaped as：Grandfather
Grandfather, father, uncle, I.There is me just necessarily to have father and grandfather, this logical data just in line with there is subordinate just necessarily to have higher level,
There is higher level not necessarily to there is the principle of subordinate to perform, due to logical data described above be tree-shaped that just father and uncle that
Bifurcated due to I and uncle not in a branch fork so not being set membership I and grandfather in a branch fork but not same
The upper two-stage at me in one-level so being grandfather grandson's relation, but I and uncle in the branch fork closed on so I am between uncle
It is close relative's relation（The priority that some things occur must have elder generation after must also having logical data equally to have with logical data above,
Have after first not necessarily having such as, I beats you, beats and formerly hinders in rear medical treatment behind wound, but what is had can also be handled with antilogical
Such as：You hit the person the medical expense after injury, and this logical data is exactly that medical expense is gone out by the producer of injury, injured manufacture
Person is whom who just has）.
(8) analysis method：It by data come these analyze datas analyzed all is that preset in advance is good that analysis of this storehouse to sentence, which is,
Analyzed by reading the data in empty file store and judged.
Such as：Ask that you have had a meal, logic analysis tree-shaped first judge database can read in empty file store on having a meal
Option inside storage data and talk with it and relation of people and then select how this exports, why then analyze will ask this
It is individual,
1. greeting or chat,
2. ask me to have a meal to deepen emotion between friend or there is certain purpose to want to bribe,
3. the too busy to get away people in some post, which needs rotating shift to have a meal, determines that who first goes to have a meal,
4. some need of work, which has enough to eat just, strength work,
Make a concrete analysis of determination methods such as：
（3,4 articles first excluded with exclusive method in several impossible examples above remove the 3rd, 4 if not colleague's can
Bar）
1. the option having a meal in empty file store is read not eat, but the people for by interpersonal relationships judge and I talks with is me
Subordinate（Probably it is intended to invite you to have a meal）Even if the invitation of subordinate should not be received so reading having a meal in empty file store
Option be do not eat must also export be eaten up or refused invite.
2. the people of question with you whether be good friend's relation and also before repeatedly together with have a meal, read the now time
Judge whether to the time having a meal if it is, the people of that question is just intended to ask you to have a meal（At this moment you are it is also contemplated that this is no
Whether please by you）,
3. if the people and your relation that can judge to talk with you by interpersonal relationships are not very ripe or unit is led（It is not clear
It is anticipated）Can with direct access inquiry other side you what can I help you, if other side be you client or you be a powerful people（Having very much can
It can be intended to invite you to have a meal）.You can be according to the regulations of unit with having a meal or do not eat come the invitation for receiving and refusing client
It may also indicate that receiving or do not receive,
If 4. a people is at table the time ask one you had a meal, then just the hasty just explanation other side that rings off is exactly one
Greeting or other side are often greeted to you in this way before more illustrates that his the words is exactly a greeting（Such as
Fruit analysis judges more than two results select one with regard to whose machine, if this storehouse be cannot analyse to ask that you have had a meal, what
Intention can also be selected according to the custom of people（Custom is exactly the word or sentence for selecting frequency most）Wait Late Stage Verification according to
People's is accustomed to the whether correct or handled using erotesis or to engineer's help of filing a request of selection）,（Should also in this storehouse
It is used to calculate plus math block）,（This storehouse carrys out the purpose of a people if analyzed but he is stored if can not saying
One people is noted inside standby option as the data for judging to need with post analysis to empty file store）.
(9) tree-shaped logic analysis judges that database can not be found at match statement to the sentence brought from input window
Reason method has two kinds（Matching is exactly a comparison process, and it is exactly to match that can compare successfully and coincide upper）：
If 1. matched and can be matched by reading the data in the tree-shaped character word stock in data filling storehouse, by reading number
Semantic analysis is carried out to the description data of word words or short sentence according to supplement storehouse the inside；As you go to Beijing, analysis method is as follows：You are
People claims noun, and it is exactly to require that you go to some address that your word, which even removes word, you go this word description data be requirement you from present
Moved to some address position（You go two words to connect address, exactly require that you go to this address behind two words to move to you）You go
The words Beijing, Beijing seeks to mobile target endpoint and calculated by analysis from a place to another place movement requirement
Distance, whether need the vehicles and preparation when going what go this target this do what, go if going out you by distance analysis
How long, when go, must understand fully can just set out, if the sentence that still can not be brought input window with this method
With success, suggesting that your word, I does not understand or sought help simultaneously to many people and help anolytic sentence.
2. handled to engineer's help of filing a request.
(10) tree-shaped logic analysis judges that database reads tree-shaped thinking framework number first to the sentence sent from input window
Progress can be passed through by judging whether that if the context mechanism for meeting tree-shaped thinking framework lane database meets according to the data in storehouse
In next step to semantic analysis, if the data for meeting tree-shaped thinking framework lane database why not must be analyzed by not meeting, point
Analysis method is such as：You is asked either with or without having a meal and your answer is for I does not sleep, question and answer be onrelevant belong to grammatically wrong sentence structure so
The data of tree-shaped thinking framework lane database are not just met, sentence above is considered as other side and do not done by the sentence for running into grammatically wrong sentence structure
Response, using hereafter as another data processing（Such as：Ask that you have had a meal, answer is what you did yesterday and go.At this time
It is adapted to not do response with sentence above is considered as other side, using hereafter as another data processing）If the above method can not
Solve, it can seek help to be formed after big data to many known people simultaneously arrives tree-shaped thinking framework number by the storage of big data result
Handled according in storehouse or to engineer's help of filing a request.（Tree-shaped logic analysis judges that database is reading tree-shaped thinking framework number
The word hereafter having during according to data in storehouse, word or sentence be accordingly to be regarded as can with the word that associates above, word or sentence, otherwise hereafter without
Word, word or sentence are accordingly to be regarded as can not be with the word that associates above, word or sentence）.Tree-shaped logic analysis judges that database judges two data
Between whether it is relevant be to be judged by reading the data of tree-shaped thinking framework lane database.
(11) tree-shaped logic analysis judges database people's name recognition method：In Part III by reading data filling storehouse
The following three kinds of methods of data in name identification database：1. it is exactly Lao Zhang that old word, which adds, small character adds that to open be exactly Xiao Zhang；2. surname
The name commonly used words in name identification database are added to be identified plus according to some position in sentence or some word, before word
Occur that the method for name is identified afterwards；3. in short some position occur name and this occur that the position of name goes out
Existing word, word, sentence are so assured that the local words that name in short occurs with word, word, sentence onrelevant below
It is exactly name）.
(12) if input window input is that data rather than sentence need the data conversion of input to divide into tree-shaped logic
Analysis judges the data that database can be handled（As input be radar feedback signal if need the signal of input to be converted into aircraft
The position at place and speed it aircraft could be monitored）.
(13) should also there be individual error correction system in this storehouse（If the sentence None- identified of input, which can not compare, needs correction process
Method has two kinds）.
Error correction method has two kinds：1 is that data can not compare when phonetically similar word or word association being run into the mistake of phonetically similar word or word
The analysis of the mistake 2 of the phonetically similar word occurred when replacing by using phonetically similar word when success can not match None- identified and just can handle writing
If method error correction method be in a word some word or word cannot with there is some word in some word or word collocation a word
Or word cannot be arranged in pairs or groups with some word or word mistake when just to analyze the words and can be replaced with what word or word be not in this
The mistake of sample, and replace erroneous words or content that the later sentence of word must be able to and talk with before matches（Such as：You remove electricity
Depending on one time.You go behind must be an address and the words you go two words to be followed by cannot be with this noun collocation of TV
So the words is grammatically wrong sentence, if the content talked with before is to go TV one to plow you on Pekinese's thing so can to make into
You go to Beijing one to plow）,（This error correction system effect is bad when with this method making machine people to use, and run into disease
It is good or handled to engineer's help of filing a request that any meaning is directly asked in reply during sentence）.
Tree-shaped logic analysis judge database in line with have sentence first handle sentence without sentence processing word without word processing word original
Then work（The sentence that the meaning of the words can find matching to the sentence of input just handles semanteme by the sentence of matching, can not find
The sentence of matching just handles semanteme by the word in sentence and carries out handling semanteme by the word in sentence if word also can not find）,（Tree
Shape logic analysis judges the analyze data that the data of lane database are made according to the ability of producer, view, idea difference
Quality, as a result also different, the sentence whether this storehouse can handle input well depends entirely on the quality of data）.
Embodiment 3
Intelligent robot manufacture method described in embodiment 1, the data of described tree-shaped thinking framework database be all it is tree-shaped,
The data of tree-shaped thinking framework database be all behind being listed in by how many kind is answered behind a word method all by row and
Every words be all synonym or sentence, near synonym or sentence put together optional one as output data but export this data must
It must can judge other of database, data filling storehouse and tree-shaped thinking framework lane database with empty file store, tree-shaped logic analysis
Data can be matched, and tree-shaped thinking framework database is made up of numerous such data（In short for behind the first DBMS
How many kind answer belongs to this structure of next DBMS and is called structure above）,（Under the data of tree-shaped thinking framework lane database
The word that text has, word or sentence be accordingly to be regarded as can with the word that associates above, word or sentence, otherwise hereafter without word, word or sentence are accordingly to be regarded as
Can not be with the word that associates above, word or sentence）；
(1) tree-shaped thinking framework library facility：If the data function in this storehouse is to define scope to subordinate's output to exceed regulation model
Enclose the phenomenon that just occurs that the language fails to express the meaning（Occurring the language fails to express the meaning, phenomenon have that two kinds of situations 1 are data deficiencies, 2 be the sentence sheet of input
Body is with regard to problematic）,（Data structure in this storehouse inherently determines scope because in short how many kind answer is all complete
Portion, which has listed, is having more the answer that a kind of answer is exactly scope as defined in grammatically wrong sentence is just above）, this storehouse major function is exactly
It is supplied to tree-shaped logic analysis to judge that database is used for detecting whether context goes beyond the scope, goes beyond the scope and just occur that word does not reach
Meaning phenomenon（The DBMS of data first in this storehouse is that the second DBMS above is the hereafter data context sheet in this storehouse
Body be exactly associate as：Ask that you have had a meal above, hereafter answer me and do not eat！This two word inherently associates）；
(2) the operation principle in tree-shaped thinking framework storehouse：The operation principle in this storehouse is to utilize the structure above of the data in this storehouse
It is supplied to tree-shaped logic analysis to judge that database judges whether input meets structure above with output（Such as ask：You have had a meal, and return
Answer eaten or do not eaten like this question sentence with to answer sentence its two sentence justice of itself be exactly to associate（This association data structure be exactly
Context mechanism）Using such associated data with regard to that can judge whether input and the sentence exported are relevant, input and output
Onrelevant just belongs to grammatically wrong sentence）（This storehouse is that the closely adopted sentence word word of all sentence of same meaning word words is put together optional one as defeated
This sentence for going out and choosing must judge database with data filling storehouse, tree-shaped logic analysis, in tree-shaped thinking framework storehouse
Other data match）；
(3) the characteristics of tree-shaped thinking framework storehouse：The characteristics of this storehouse be run into why, word as whether what's the matter, be not or not etc.,
Word or sentence can be used only in some data hereafter the inside use, it is not possible to as independent data use still why, what's the matter
The word Deng as can be used as independent data and can associated with other related datas with after the collocation of other words（This storehouse
In data run into name and include you, if I, he, man, the word such as Ms all for it is empty with bracket come instead of but inside bracket
Insert the first person, the second person, the third person and the first person, the second person, the third person point to man or woman or
It is that DBMS structure is also different at present for the name of the name woman of man）.
Embodiment 4
Intelligent robot manufacture method described in embodiment 1, the data of described empty file store are made up of two parts, this two parts
Data are dummy status, and empty file store is exactly the memory system of robot, and the data of empty file store divide two parts, a portion
It is the result that record includes exporting after analysis judges on the information of robot oneself, another part is the people that record is talked with it
Information include the people that talks with it what is said or talked about；
Two parts data of empty file store are tree；
（Data such as the inside have the option had a meal or do not had a meal option but had a meal can be recorded behind option of having a meal
Also have that eats what is and to eat more and few option etc. below）；
The data of empty file store divide two parts, and a portion is that record is sentenced on the information of robot oneself including logic analysis
Have no progeny result, another part is the information for the people that record is talked with it.
(1) the function of empty file store：This library facility is to judge database to its storage from tree-shaped logic analysis or read number
According to,（This storehouse is also required to the original words of input window input to carry out storage as session log）；
(2) the characteristics of empty file store：Data in this storehouse are at some letters of the state only with respect to robot oneself of sky
Breath is not just setting in advance for sky（These information can also be not configured to sky with whose meaning setting can also be by robot
It is oneself setting work oneself to set still robot oneself setting must to have its ability of the data of correlation oneself to complete）, this
Data options inside individual empty file store are all to record some as whose work the fat or thin home background friend relative of the height of people have
Many data options such as unit educational background, just stored, deposited by the content of data options here when having declarative sentence input
These data are not just empty after storage, and robot he can also be touched everyone all establish as one
Empty file store, there is the data machine talent conference in empty file store understands that he is touched everyone,（This storehouse is with regard to suitable
In the form of a very full resume, the inside records the home background of a people and the personal fat or thin grade of height works as us
Seeing these contents, we will appreciate that this people, and robot is similarly read after these data it will also be appreciated that a people, this
Individual storehouse is exactly the memory system of robot（This storehouse should also have note alternate item to be used for storing the view, idea etc. to a people
Information）,（Empty file store should also have unfinished dialogue the Save option, for storing unfinished dialogue when talking with next time
Use（Such as：Someone and robot say that you guess that whom I saw and then just rang off just now, when you see robot in second day again
, robot can ask you, and whom you saw yesterday）,（This storehouse must be recorded in record data the time otherwise he will be hard to tell
It is word or present word before 10 years）,（Call the data in supplement storehouse must be supplement when data deficiencies in this storehouse
Data duplication in storehouse could use to empty file store the inside）.
Embodiment 5
Intelligent robot manufacture method described in embodiment 3, described data filling storehouse are divided into three parts：
Part I carries out supplementary data to empty file store and used, when empty file store data deficiencies standby it supplement；
Part II carries out supplementary data to tree-shaped thinking framework storehouse and used, the standby when tree-shaped thinking framework storehouse data deficiencies
It is supplemented；
Part III judges that tree-shaped logic analysis database carries out supplementary data and used, when tree-shaped logic analysis judges data
During the data deficiencies of storehouse standby it supplement.
Part III must have tree-shaped character word stock and name an identification database（Tree-shaped character word stock also includes short sentence, this
Tree-shaped character word stock is also synonymous near synonym, and synonymous nearly adopted word is put together to be handled by the same meaning）：
(1) the tree-shaped character word stock data structure in this storehouse is such as：The some words of certain class or word can some with certain class words or word be connected this
A bit connected word or word again can some with certain class words or word be connected the composition tree shaped data of one-level one-level like this, these data
Untill one is connected directly to be added（It is the same such as just as our word combining and sentence making：Your word can become reconciled word, word, remove word, arrive word etc.
Word or word are connected, your two words can also with go word be connected one be connected directly to connect untill, having to the word in character word stock also
Or the description data of word）, this storehouse should also press the words in this storehouse noun, verb, adjective, number, and measure word etc. enters
Row classification（More multiclass can also be separated below noun verb-adjective etc. such as：It can be divided under noun：Place name, vegetables name, business
Name of an article word, commodity noun with the classes such as ignition distributor and commodity, the purpose of classification are searched for the ease of quick again）：This word
Also the word in this storehouse and word should be described for dictionary（Description data should include word, the usage of word, as face like peach blossom this
Individual word is most to be used in the appearance for describing people）：Such as the description of commodity noun mobile phone, its volume weight belongs to which kind of purposes, usage
Deng description such as verb run description run speed run rule run method description,（Such as going the description data of word more, going
The characteristics of being and removing word to another place movement from a place, every must all have a people to claim noun by the sentence for going word to form
Point to a people and it is every must all be had by the sentence for going word form an address noun or sensing one address）, have
The tree-shaped logic analysis of these data judges that database can handle the implication of the sentence of these words or word composition.
(2) name identification database：
Data in this storehouse should have the conventional word and word of One Hundred Family Names storehouse and name.
（If tree-shaped logic analysis judges that database oneself can analyze new logic analysis data and also be stored in this
In data filling storehouse secondary analysis are avoided the need for as supplementary data）.
（The time must be recorded when all data are recorded in this method, also robot must must know and his dialogue
People be who it could handle the sentence of this people input and otherwise handled by stranger）.
（Not necessarily handy word can also use other forms to the date storage method of the present invention as data storage method
No matter using what form as its operation principle of the storage method of data or as）.
Certainly, described above is not limitation of the present invention, and the present invention is also not limited to the example above, this technology neck
The variations, modifications, additions or substitutions that the technical staff in domain is made in the essential scope of the present invention, it should also belong to the present invention's
Protection domain.
Claims (5)
1. a kind of intelligent robot manufacture method, it is characterized in that：Realized by following steps：
Step 1：Robot oneself first using the sentence inputted from input window give first sky file store carry out storage as pair
Also give tree-shaped logic analysis judges that database is analyzed and judges that data can be read in analysis and judgement to words record simultaneously
Supplement storehouse in data assist analysis then by the result analyzed and judged export to output window complete output while
The result analyzed and judged can be given to sky file store and carry out storage as session log；
Step 2：Tree-shaped logic analysis judges that database can need to read tree-shaped thinking framework in analysis and judgement according to data
The data of lane database assist analysis or to tree-shaped thinking framework data database storings；
Step 3：Tree-shaped logic analysis judges that database can need to read empty file store data in analysis and judgement according to data
Assist analysis or to neutral gear case library storage data；
Step 4：When tree-shaped logic analysis judges database, tree-shaped thinking framework database, read during empty file store data deficiencies
Data in data filling storehouse can also be judged that database is mended to data as supplementary data or as needed from tree-shaped logic analysis
Fill library storage data.
2. intelligent robot manufacture method according to claim 1, it is characterized in that：Described tree-shaped logic analysis judges number
Data according to storehouse are all tree-shaped and are that the sentence of same meaning, nearly adopted sentence are put together optional one as output data, but exported
This data must be able to judge data with empty file store, tree-shaped thinking framework database, data filling storehouse and tree-shaped logic analysis
Other data in storehouse can match, and tree-shaped logic analysis judges the data of database are exactly in short how many kind implication and right
The words needs the data analyzed all by behind the coming of row one-level one-level, tree-shaped logic analysis judge database have it is numerous this
The data composition of sample, tree-shaped logic analysis judge that the data of lane database should include to semantic analysis data and to cycle, body
Product, speed, weight, volume, volume, pressure, inertia force, area, height, interpersonal relation, between object and object
Relation, the characteristics of motion, movement technique, object structures, the processing data of mobile changing rule method between position and position.
3. intelligent robot manufacture method according to claim 1, it is characterized in that：Described tree-shaped thinking framework database
Data be all it is tree-shaped, the data of tree-shaped thinking framework database be all by how many kind is answered behind a word method all
By row, be listed in behind and also every words be all that the sentence of same meaning, nearly adopted sentence are put together optional one as output data but output
This data must be able to judge database, data filling storehouse and tree-shaped thinking framework data with empty file store, tree-shaped logic analysis
Other data in storehouse can be matched, and tree-shaped thinking framework database is made up of numerous such data.
4. intelligent robot manufacture method according to claim 1, it is characterized in that：The data of described empty file store are by two
Part is formed, and this two parts data is dummy status, and empty file store is exactly the memory system of robot, and the data of empty file store are divided
Two parts, a portion are that record includes analyzing the result exported after judgement, another part on the information of robot oneself
Being the information of people talked with it of record, what is said or talked about including the people that talks with it.
5. intelligent robot manufacture method according to claim 3, it is characterized in that：Data filling storehouse is divided into three parts：
Part I carries out supplementary data to empty file store and used, when empty file store data deficiencies standby it supplement；
Part II to tree-shaped thinking frame data frame storehouse carry out supplementary data use, when tree-shaped thinking frame data frame storehouse data not
When sufficient standby it supplement；
Part III judges that tree-shaped logic analysis database carries out supplementary data and used, when tree-shaped logic analysis judges data
During the data deficiencies of storehouse standby it supplement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710846773.5A CN107633052A (en) | 2017-09-19 | 2017-09-19 | Intelligent robot manufacture method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710846773.5A CN107633052A (en) | 2017-09-19 | 2017-09-19 | Intelligent robot manufacture method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107633052A true CN107633052A (en) | 2018-01-26 |
Family
ID=61103318
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710846773.5A Pending CN107633052A (en) | 2017-09-19 | 2017-09-19 | Intelligent robot manufacture method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107633052A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102868695A (en) * | 2012-09-18 | 2013-01-09 | 天格科技（杭州）有限公司 | Conversation tree-based intelligent online customer service method and system |
| CN105528143A (en) * | 2006-04-21 | 2016-04-27 | 泰吉克通讯股份有限公司 | Contextual prediction of user words and user actions |
| US9473447B1 (en) * | 2014-10-27 | 2016-10-18 | Rushline, LLC | Systems and methods for enabling dialog amongst different participant groups |
| CN106055537A (en) * | 2016-05-23 | 2016-10-26 | 王立山 | Natural language machine recognition method and system |
-
2017
- 2017-09-19 CN CN201710846773.5A patent/CN107633052A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105528143A (en) * | 2006-04-21 | 2016-04-27 | 泰吉克通讯股份有限公司 | Contextual prediction of user words and user actions |
| CN102868695A (en) * | 2012-09-18 | 2013-01-09 | 天格科技（杭州）有限公司 | Conversation tree-based intelligent online customer service method and system |
| US9473447B1 (en) * | 2014-10-27 | 2016-10-18 | Rushline, LLC | Systems and methods for enabling dialog amongst different participant groups |
| CN106055537A (en) * | 2016-05-23 | 2016-10-26 | 王立山 | Natural language machine recognition method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106295792B (en) | Dialogue data interaction processing method and device based on multi-model output | |
| CN109710772B (en) | Question-answer base knowledge management system based on deep learning and implementation method thereof | |
| EP3529717B1 (en) | Multi domain real-time question answering system | |
| EP3804915A1 (en) | Conversational systems and methods for robotic task identification using natural language | |
| CN110162611A (en) | A kind of intelligent customer service answer method and system | |
| US20170337180A1 (en) | Recognition method and system of natural language for machine thinking | |
| CN109918489A (en) | A kind of knowledge question answering method and system of more strategy fusions | |
| CN109344237A (en) | A kind of method and device of the information processing for human-computer interaction | |
| CN105589844A (en) | Missing semantic supplementing method for multi-round question-answering system | |
| CN109492101A (en) | File classification method, system and medium based on label information and text feature | |
| US11238075B1 (en) | Systems and methods for providing inquiry responses using linguistics and machine learning | |
| CN110175229A (en) | A kind of method and system carrying out online training based on natural language | |
| CN107341251A (en) | A kind of extraction and the processing method of medical folk prescription and keyword | |
| CN113987151A (en) | Intelligent customer service reply method and device | |
| CN110232180A (en) | A kind of automatic proposition method and system towards classic poetry assessment | |
| CN111831801A (en) | A man-machine dialogue method and system | |
| GB2583196A (en) | Search engine optimization techniques | |
| KR101333485B1 (en) | Method for constructing named entities using online encyclopedia and apparatus for performing the same | |
| CN109712645A (en) | Autonomous phone system and autonomous call method | |
| CN107633052A (en) | Intelligent robot manufacture method | |
| Bromberger | Science and the Forms of Ignorance | |
| CN109165281A (en) | A kind of shared legal advice system of multirobot terminal based on semantic net | |
| CN110414465B (en) | Emotion analysis method for video communication | |
| Nirenburg et al. | Slashing metaphor with Occam’s Razor | |
| Barco et al. | Sentiment Analysis in Spanish Reviews: Dataket Submission on REST-Mex 2022. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |